h1aji / chmsee

CHM viewer
GNU General Public License v2.0
1 stars 1 forks source link

speed issures using a relatively big file #129

Closed GoogleCodeExporter closed 8 years ago

GoogleCodeExporter commented 8 years ago
1.Opening a relatively big file seems takes too much time, the problem also 
exists when using xchm.I wonder whether chmsee can build an index file like 
.chw in windows or simply read the .chw file already existing.

2.Another problem is that when using "index" tab for searching it becomes 
unbearably slow.Xchm doesn't have this problem since it only shows the first 
choice.

Thanks for the wonderful chm reader!

Hongxu Chen

Original issue reported on code.google.com by leftcopy...@gmail.com on 24 Jul 2011 at 2:56

GoogleCodeExporter commented 8 years ago
1. Yes, opening a new chm file is slow, but in the second time, this procedure 
will be fast.

2. I've tested this feature on chmsee 2.0 with "php manual" - it has a 2M 
index(hhk) file - and didn't feel any pause. Could you tell me which chmsee 
version and which chm file you are testing?

Original comment by jungl...@gmail.com on 24 Jul 2011 at 3:57

GoogleCodeExporter commented 8 years ago
The version is 1.3.0 and I'm opening a chm file with the size of about 40M. 
There seems no hhk file in $HOME/.cache/chmsee or the current file 
directory.Each time opening the chm file costs a long time.I'm just an ordinary 
user of ubuntu and don't want to bother to compile the src package.Many thanks.

Original comment by leftcopy...@gmail.com on 24 Jul 2011 at 10:41

GoogleCodeExporter commented 8 years ago
Is there a link for this 40M or similar chm files? 
Let me test them to see whether the developing chmsee has the same problem or 
not. 

Original comment by jungl...@gmail.com on 27 Jul 2011 at 2:45

GoogleCodeExporter commented 8 years ago
It's about 50M .Many thanks.

Original comment by leftcopy...@gmail.com on 2 Aug 2011 at 1:58

GoogleCodeExporter commented 8 years ago
I tried to open a very large file in Ubuntu 12.04 with chmsee, I mean a 1.8 Gb 
file (a medical text book) and the program got stuck. The same happened in 
different 32 bit and 64 bit computers. That does not happen with Kchmviewer, a 
very good program by he way, even with the overhead of loading all the KDE 
libs. The bug is very reproducible.

Original comment by danylis...@gmail.com on 5 Jul 2012 at 5:16

GoogleCodeExporter commented 8 years ago
The chmsee way is to unzip chm file to local file system, then use mozilla 
gecko rendering these htmls. So, when you open a huge chm file, it will spend 
time on extracting and need more hard drive space. This is chmsee's 
disadvantage.

Original comment by jungl...@gmail.com on 7 Jul 2012 at 2:07

GoogleCodeExporter commented 8 years ago
Couldn't chmsee extract the index and work out which files to extract 
on-the-fly while viewing the chm file?
Like this: user opens chm file, chmsee extracts index, user selects topic, 
chmsee extracts relevant files to cache folder. This way the cache is built in 
a more convenient way, and won't take up 200MB of your disk if you open a chm 
file of that size.

I know this involves some HTML parsing and whatnot, but I think it would be a 
good option. ChmFox (a firefox extension) reads chm files too, and opens them 
fast. I'd check out how it does that, but I'm currently out of time.

Original comment by a.pedro....@gmail.com on 14 Sep 2012 at 1:52

GoogleCodeExporter commented 8 years ago
I agree, extracting on demand is a better solution than the way chmsee 
currently using.

If converting the method, chmsee must respond every request for every object on 
a html page and add some hooks on xulrunner to interrupt xulrunner processing. 
I think these features are a http server's behavior and chmsee as a desktop 
program should not be involved. 

Another reason chmsee choosing extracting before demand is storage is very 
cheap nowadays, so wastes some disk space is acceptable for desktop user.

Original comment by jungl...@gmail.com on 14 Sep 2012 at 3:22

GoogleCodeExporter commented 8 years ago
Clear all pre-2.0 issues

Original comment by jungl...@gmail.com on 18 Jan 2013 at 6:10